Hiya,
I agree with Tim that THERMOCALC will provide good results from the GADS
barometer in average P mode.
Another option is to run your input file (obtained from AX) in mode
3:calculations on all reactions between end-members, when prompted include
only the end members for the reaction you're interested in ie. di an gr py
q, TC will then plot that reaction for you, over the PT window you
specify.
Additionally if you include almandine and hedenbergite in the end members
you will also get the Fe version of the baromter plus the alm+di=py+hed
reaction (gt-cpx thermometer) TC then nicely calculates the intersection
of these three reactions for you, giving you a PT point plus errors on the
estimate.
At that point if you want to you could even calculate the error elipse
around the intersection via Powell and Holland (1988).
Cheers,
Jon
PS. For anybody who is interested I've attached a quickly made up input
file for TC plus the logfile and output files for an example calculation.
PPS. If there are any problems with this please be aware that I've been
writing up for a long time and any criticism-induced frustration will
impact directly upon my office mate, who is a fragile soul. So think of
him before you reply :)
|